avformat: add subtitle support in master playlist m3u8

Test with the following command for the webvtt subtitle:
$ ./ffmpeg -y -i input_with_subtitle.mkv \
 -b:v:0 5250k -c:v h264 -pix_fmt yuv420p -profile:v main -level 4.1 \
 -b:a:0 256k \
 -c:s webvtt -c:a mp2 -ar 48000 -ac 2 -map 0:v -map 0:a:0 -map 0:s:0 \
 -f hls -var_stream_map "v:0,a:0,s:0,sgroup:subtitle" \
 -master_pl_name master.m3u8 -t 300 -hls_time 10 -hls_init_time 4 -hls_list_size \
 10 -master_pl_publish_rate 10  -hls_flags \
 delete_segments+discont_start+split_by_time ./tmp/video.m3u8

Check the master m3u8:
$ cat tmp/master.m3u8
#EXTM3U
#EXT-X-VERSION:3
#EXT-X-MEDIA:TYPE=SUBTITLES,GROUP-ID="subtitle",NAME="subtitle_0",DEFAULT=YES,URI="video_vtt.m3u8"
#EXT-X-STREAM-INF:BANDWIDTH=6056600,RESOLUTION=1280x720,CODECS="avc1.4d4829,mp4a.40.33",SUBTITLES="subtitle"
video.m3u8

Check the result by convert to mkv:
$ ./ffmpeg -strict experimental -i ./tmp/master.m3u8 -c:v copy -c:a mp2 -c:s srt ./test.mkv

Signed-off-by: Limin Wang <lance.lmwang@gmail.com>

5 files changed, 59 insertions, 5 deletions

diff --git a/doc/muxers.texi b/doc/muxers.texi
index 3be1c89416..4b6ffaaf58 100644
--- a/doc/muxers.texi
+++ b/doc/muxers.texi
@@ -1062,6 +1062,21 @@ have and language is named ENG, the other audio language is named CHN.
 
 By default, a single hls variant containing all the encoded streams is created.
 
+@example
+ffmpeg -y -i input_with_subtitle.mkv \
+ -b:v:0 5250k -c:v h264 -pix_fmt yuv420p -profile:v main -level 4.1 \
+ -b:a:0 256k \
+ -c:s webvtt -c:a mp2 -ar 48000 -ac 2 -map 0:v -map 0:a:0 -map 0:s:0 \
+ -f hls -var_stream_map "v:0,a:0,s:0,sgroup:subtitle" \
+ -master_pl_name master.m3u8 -t 300 -hls_time 10 -hls_init_time 4 -hls_list_size \
+ 10 -master_pl_publish_rate 10  -hls_flags \
+ delete_segments+discont_start+split_by_time ./tmp/video.m3u8
+@end example
+
+This example adds @code{#EXT-X-MEDIA} tag with @code{TYPE=SUBTITLES} in
+the master playlist with webvtt subtitle group name 'subtitle'. Please make sure
+the input file has one text subtitle stream at least.
+
 @item cc_stream_map
 Map string which specifies different closed captions groups and their
 attributes. The closed captions stream groups are separated by space.
diff --git a/libavformat/dashenc.c b/libavformat/dashenc.c
index 94d463972a..d1fe90b00c 100644
--- a/libavformat/dashenc.c
+++ b/libavformat/dashenc.c
@@ -1311,7 +1311,7 @@ static int write_manifest(AVFormatContext *s, int final)
             get_hls_playlist_name(playlist_file, sizeof(playlist_file), NULL, i);
             ff_hls_write_stream_info(st, c->m3u8_out, stream_bitrate,
                                      playlist_file, agroup,
-                                     codec_str_ptr, NULL);
+                                     codec_str_ptr, NULL, NULL);
         }
         dashenc_io_close(s, &c->m3u8_out, temp_filename);
         if (use_rename)
diff --git a/libavformat/hlsenc.c b/libavformat/hlsenc.c
index ae5eae4a7d..fda9431edc 100644
--- a/libavformat/hlsenc.c
+++ b/libavformat/hlsenc.c
@@ -164,6 +164,7 @@ typedef struct VariantStream {
     int is_default; /* default status of audio group */
     char *language; /* audio lauguage name */
     char *agroup; /* audio group name */
+    char *sgroup; /* subtitle group name */
     char *ccgroup; /* closed caption group name */
     char *baseurl;
     char *varname; // variant name
@@ -1285,7 +1286,9 @@ static int create_master_playlist(AVFormatContext *s,
     unsigned int i, j;
     int ret, bandwidth;
     const char *m3u8_rel_name = NULL;
+    const char *vtt_m3u8_rel_name = NULL;
     char *ccgroup;
+    char *sgroup = NULL;
     ClosedCaptionsStream *ccs;
     const char *proto = avio_find_protocol_name(hls->master_m3u8_url);
     int is_file_proto = proto && !strcmp(proto, "file");
@@ -1408,13 +1411,24 @@ static int create_master_playlist(AVFormatContext *s,
                         vs->ccgroup);
         }
 
+        if (vid_st && vs->sgroup) {
+            sgroup = vs->sgroup;
+            vtt_m3u8_rel_name = get_relative_url(hls->master_m3u8_url, vs->vtt_m3u8_name);
+            if (!vtt_m3u8_rel_name) {
+                av_log(s, AV_LOG_WARNING, "Unable to find relative subtitle URL\n");
+                break;
+            }
+
+            ff_hls_write_subtitle_rendition(hls->m3u8_out, sgroup, vtt_m3u8_rel_name, vs->language, i, hls->has_default_key ? vs->is_default : 1);
+        }
+
         if (!hls->has_default_key || !hls->has_video_m3u8) {
             ff_hls_write_stream_info(vid_st, hls->m3u8_out, bandwidth, m3u8_rel_name,
-                    aud_st ? vs->agroup : NULL, vs->codec_attr, ccgroup);
+                    aud_st ? vs->agroup : NULL, vs->codec_attr, ccgroup, sgroup);
         } else {
             if (vid_st) {
                 ff_hls_write_stream_info(vid_st, hls->m3u8_out, bandwidth, m3u8_rel_name,
-                                         aud_st ? vs->agroup : NULL, vs->codec_attr, ccgroup);
+                                         aud_st ? vs->agroup : NULL, vs->codec_attr, ccgroup, sgroup);
             }
         }
     }
@@ -1889,6 +1903,7 @@ static int parse_variant_stream_mapstring(AVFormatContext *s)
      * practical usage)
      *
      * agroup: is key to specify audio group. A string can be given as value.
+     * sgroup: is key to specify subtitle group. A string can be given as value.
      */
     p = av_strdup(hls->var_stream_map);
     if (!p)
@@ -1956,6 +1971,12 @@ static int parse_variant_stream_mapstring(AVFormatContext *s)
                 if (!vs->agroup)
                     return AVERROR(ENOMEM);
                 continue;
+            } else if (av_strstart(keyval, "sgroup:", &val)) {
+                av_free(vs->sgroup);
+                vs->sgroup = av_strdup(val);
+                if (!vs->sgroup)
+                    return AVERROR(ENOMEM);
+                continue;
             } else if (av_strstart(keyval, "ccgroup:", &val)) {
                 av_free(vs->ccgroup);
                 vs->ccgroup = av_strdup(val);
@@ -2512,6 +2533,7 @@ static void hls_free_variant_streams(struct HLSContext *hls)
         av_freep(&vs->m3u8_name);
         av_freep(&vs->streams);
         av_freep(&vs->agroup);
+        av_freep(&vs->sgroup);
         av_freep(&vs->language);
         av_freep(&vs->ccgroup);
         av_freep(&vs->baseurl);
diff --git a/libavformat/hlsplaylist.c b/libavformat/hlsplaylist.c
index 56244496c0..43f9d281ba 100644
--- a/libavformat/hlsplaylist.c
+++ b/libavformat/hlsplaylist.c
@@ -48,9 +48,22 @@ void ff_hls_write_audio_rendition(AVIOContext *out, char *agroup,
     avio_printf(out, "URI=\"%s\"\n", filename);
 }
 
+void ff_hls_write_subtitle_rendition(AVIOContext *out, char *sgroup,
+                                  const char *filename, char *language, int name_id, int is_default) {
+    if (!out || !filename)
+        return;
+
+    avio_printf(out, "#EXT-X-MEDIA:TYPE=SUBTITLES,GROUP-ID=\"%s\"", sgroup);
+    avio_printf(out, ",NAME=\"subtitle_%d\",DEFAULT=%s,", name_id, is_default ? "YES" : "NO");
+    if (language) {
+        avio_printf(out, "LANGUAGE=\"%s\",", language);
+    }
+    avio_printf(out, "URI=\"%s\"\n", filename);
+}
+
 void ff_hls_write_stream_info(AVStream *st, AVIOContext *out,
                               int bandwidth, const char *filename, char *agroup,
-                              char *codecs, char *ccgroup) {
+                              char *codecs, char *ccgroup, char *sgroup) {
 
     if (!out || !filename)
         return;
@@ -71,6 +84,8 @@ void ff_hls_write_stream_info(AVStream *st, AVIOContext *out,
         avio_printf(out, ",AUDIO=\"group_%s\"", agroup);
     if (ccgroup && ccgroup[0])
         avio_printf(out, ",CLOSED-CAPTIONS=\"%s\"", ccgroup);
+    if (sgroup && sgroup[0])
+        avio_printf(out, ",SUBTITLES=\"%s\"", sgroup);
     avio_printf(out, "\n%s\n\n", filename);
 }
 
diff --git a/libavformat/hlsplaylist.h b/libavformat/hlsplaylist.h
index a8d29d62d3..a124bdcffb 100644
--- a/libavformat/hlsplaylist.h
+++ b/libavformat/hlsplaylist.h
@@ -39,9 +39,11 @@ typedef enum {
 void ff_hls_write_playlist_version(AVIOContext *out, int version);
 void ff_hls_write_audio_rendition(AVIOContext *out, char *agroup,
                                   const char *filename, char *language, int name_id, int is_default);
+void ff_hls_write_subtitle_rendition(AVIOContext *out, char *sgroup,
+                                  const char *filename, char *language, int name_id, int is_default);
 void ff_hls_write_stream_info(AVStream *st, AVIOContext *out,
                               int bandwidth, const char *filename, char *agroup,
-                              char *codecs, char *ccgroup);
+                              char *codecs, char *ccgroup, char *sgroup);
 void ff_hls_write_playlist_header(AVIOContext *out, int version, int allowcache,
                                   int target_duration, int64_t sequence,
                                   uint32_t playlist_type, int iframe_mode);